A Passive Flux Denuder for Evaluating Emissions of Ammonia at a Dairy Farm
Dennis R. Fitz, College of Engineering-Center for Environmental Research and Technology, University of California, Riverside, CA 92521 John T. Pisano, College of Engineering-Center for Environmental Research and Technology, University of California, Riverside, CA 92521 Dave Goorahoo, Center for Irrigation Technology, California State University, Fresno, CA 93740 Charles F. Krauter, Department of Plant Science, California State University, Fresno, CA 93740 Irina L. Malkina, College of Engineering-Center for Environmental Research and Technology, University of California, Riverside, CA 92521
IMPLICATIONS
The measurement of ammonia emission rates from fugitive sources traditionally has required expensive air sampling methods to measure concentration and wind speed. The passive flux denuder has been shown to be a simple and effective method to measure these emissions rates.
The sampling equipment is inexpensive, no electricity or batteries are needed to collect samples, and the sampling can be done with inexperienced personnel. This approach will facilitate local agencies in measuring emission rates from such fugitive emission sources. The affected industries can also easily monitor the effectiveness of control strategies to mitigate these sources.
1 ABSTRACT
There is significant interest in developing economical methods to quantify ambient
concentrations of gaseous pollutants. Passive samplers can provide an inexpensive alternative to
direct sampling. Conventional denuder technology has been developed to measure semi-volatile
pollutants, but active sampling inherently requires electrical power and pumps. The idea of
comparing active sampling with a passive configuration, specifically for ammonia flux, came
from the need to estimate ammonia emissions from area sources in California. Passive and active
denuders were collocated at a dairy farm at the California State University, Fresno, Agricultural
Research Facility. In this project, ammonia measurements were made for the dairy farm lagoon
as it underwent acidification, and for the dairy farm as a whole. Comparisons were made of the
flux measurements obtained directly from the passive flux denuder and those calculated from an
active filterpack sampler and wind velocity. The results show significant correlation between the two methods and invite further investigation into characterization of the passive flux denuder response. With further evaluation of this technique, it is possible that a larger inventory base for
ambient ammonia emissions, can be developed more economically than by using active
samplers.
INTRODUCTION
Ammonia (NH3) is an important trace constituent of the gaseous composition that make up the
lower troposphere. It is the dominant gaseous base and is responsible for determining the level of
atmospheric acidity1. The role that ammonia plays in neutralizing acidic aerosols has led to many
2,3 studies concerning health effects of atmospheric aerosols . Higher local emissions of NH3 result
in the enhancement of aerosol formation that ultimately result in the following:
2
1) Formation of ammonium nitrate NH4NO3 by reacting with HNO3.
2) Formation of sulfate as the increase in pH enhances the rate of oxidation of dissolved
sulfur dioxide by ozone4.
3) Formation of either ammonium sulfate (NH4HSO4), or ammonium bisulfate ((NH4)2SO4)
by neutralization of sulfuric acid5.
4) Formation of ammonium chloride by reacting with HCl6.
7 NH3 has a lifetime usually of 1-5 days but once transformed to ammonium aerosol has a
lifetime that increases by at least a factor of three8. This allows for more widespread pertubation of soil acidity as well as direct deposition to plants resulting in excess nitrogen. Excess nitrogen is known to enhance the replacement of slow-growing plant species with fast-growing grass species9, causing adverse effects on many types of vegetation. In coastal areas excess nitrogen alters the growing frequencies of both toxic and non-toxic phytoplankton, seriously affecting this delicate ecosystem10. Increased production of ammonium aerosol also enhances light scattering, resulting in visibility degradation11.
It is known that the largest contributor of ammonia to the global budget is from domestic animal waste, as those from dairy, cattle, and swine farms12. Although there is now increased concern about NH3 emissions from mobile sources, there is still a strong need to evaluate domestic waste sources. California is ranked highest in milk production, producing 26 billion pounds of milk and cheese13. In 1998, livestock cash receipts totaled $6.85 billion, due mainly to increased milk and cream sales13. While the growth of this industry has resulted in significant economic returns for the region, there is the issue of effective manure management. As much as
40% of feed dry matter fed to the animals can be excreted in the manure, requiring a significant
3 amount of labor and capital investment for manure management and disposal14. In dairy
operations, manure is commonly handled as an effluent stream of liquid or slurry manure by means of a hydraulic flushing – lagoon storage – irrigation system. A major problem associated with the manure management is the loss of ammonia produced during the decomposition of manure in storage lagoons.
Between 80 and 200 different gases have been identified during the decomposition of
15 manure . Of these gases, NH3 is of primary concern because of: (a) the health issues related to
particulate matter (PM) formation and (b) the loss of important plant nutrients from the manure.
+ For example, NH3 is volatilized from lagoons, whereby the valuable ammonium ion (NH4 ) is
16 lost as NH3 . Volatilized NH3 can react in the atmosphere to produce ammonium nitrate or ammonium sulfate and thereby contribute to airborne particulate matter (PM). The California Air
Resources Board (ARB) has developed preliminary emissions inventories for NH3 from most potential sources in the state17, all of which were reported with high initial uncertainties for source contributions. In the urban environment of California, ammonium nitrate accounts for between 30 and 60 percent of the fine aerosol mass. The Federal government and the State of
California have established ambient air quality standards for particulate matter with aerodynamic diameters of both 10 microns or less (PM10) and 2.5 microns or less (PM2.5), which invites further concern of the overall ammonia contribution to the fine aerosol mass.
As a result of the health, environmental, economic and regulatory concerns, there is a need to quantify NH3 emissions at dairies. Since NH3 is one of the by-products of microbial degradation of manure and organic matter, a major location for this gas emission in a dairy operation is the lagoon, where the manure is stored and can be subjected to aerobic or anaerobic conditions18. In addition, gaseous emissions from these lagoons are influenced by climatic factors such as wind
4 speed, temperature, humidity and precipitation15,19, as well as, the pH of the dairy effluent18. For
example, acidification of cattle slurry applied to grasslands has been used to reduce NH3
20 emission rates . The objective of the current study was to compare the NH3 fluxes determined by passive flux denuders with those determined by an active sampling technique.
EXPERIMENTAL
Active Sampling
Filter packs were used to collect ammonia gas21. These consisted of a 47mm diameter Teflon front filter to remove particulate ammonium followed by a glass fiber filter treated with citric acid (5% in 95% ethanol and dried) to adsorb gaseous ammonia. A commercially available 12- volt air sampling pump was used to pull air through the filter pack at a nominal flow rate of 4 liters per minute. The treated filters were extracted with distilled water, reacted with Nessler's
Reagent, and the absorbance was read on a spectrophotometer at 425nm. The amount of ammonia on the filter was determined in mg NH3, after correction for blank filters, using a
+ calibration curve relating absorbance to ppm NH4 -N in solution. The colorimetric procedure involving the use of the spectrophotometer is the method commonly used among dairy operators in the region for determining ammonium concentrations in the dairy effluent used for irrigating fields22.
The ammonia flux was determined by multiplying the denuder concentrations by the average wind speed at a given location. Anemometers were collocated with filter packs at 1, 2, and 5 meters above the soil surface to measure wind speed and direction. Finally, to be consistent
5 -2 -1 with flux data reported from the passive denuders (described later) the µg N-NH3 m s values
-2 -1 converted to ng N-NH3 cm min .
Passive Sampling
Fabric Denuders. Diffusion denuders have been developed to sample air pollutants that are semi-volatile such as ammonium nitrate, which is a solid particle in equilibrium with ammonia and nitric acid vapor. These devices selectively and irreversibly remove the gas phase components while allowing particles to pass unattenuated. Diffusion denuders were first constructed as tubing bundles23, progressed to annular geometries24, and then to honeycomb structures25. Fitz and Motallebi recently reported the successful use of fabric denuders to selectively remove nitric acid26. The denuder sampling approach they proposed is based on diffusion batteries to remove very fine particles. The use of cloth fabric as the denuder substrate is analogous to the wire screen denuders for collecting very fine particles in diffusion batteries.
Cloth, with a typical thread size of 100 µm spaced on centers of 250 µm, leaving an open grid of
150 µm (typical dimensions for cloth), has been shown to efficiently remove gaseous species when coated with a suitable adsorbent. To estimate removal efficiency, comparisons to theoretically calculated values using the theory developed for wire diffusion screens were conducted. Fabric denuders were first evaluated by looking at the removal efficiency of HNO3 by
26 Na2CO3 coated examples . Fitz and Motallebi observed that they could estimate denuder
efficiency by estimating the fractional penetration of the gas and assume that the denuder is
treated with a chemical that quantitatively removes a target gas, when the gas in question comes
in contact with the denuder surface26. They concluded that two denuders made of a cotton fabric,
47 mm in diameter, were required in series to ensure a collection of over 95% of the HNO3 acid in ambient air when sampling up to 10 L/min.
6
Ammonia Removal by Fabric Denuders. To build upon this work and possibly apply this
type of denuder to ammonia measurements, work was done to investigate the efficiency of
ammonia removal by using phosphoric acid as the coating material27. Denuders were evaluated
in the laboratory by passing known concentrations of ammonia (NH3) through them at various
temperatures and humidities. Ammonia concentrations were generated using commercial
permeation tubes with specified emission rates. Gas mixtures in the range of 10-40 ppb were generated at relative humidities of 10-80%. Concentration measurements, used to determine penetration, were made before and after the denuder. The concentration of ammonia was monitored by reducing NH3 to NO by passing the sample through a high temperature stainless steel converter and then measuring the NO concentration with a commercial chemiluminescent
NO analyzer. Denuders coated with 2% phosphoric acid and with 9% phosphoric acid were tested for collection of ammonia. Table 1 shows a summary of the results of testing phosphoric acid coated fabric denuders. Overall, the collection efficiency of a single 9% phosphoric acid denuder is better than 90%, and two denuders used in series would have better than 99% collection. This denuder is therefore suitable for field sampling of ambient ammonia.
Construction of a Passive Sampler Using Fabric Denuders
The passive flux sampler design is based on the approach used by Schjoerring28, but, instead of using denuder tubes, the design was modified to use 47mm fabric denuders. Like filter types of active samplers, an open-face Savillex Teflon filter holder (Minnetonka, MN) is used to locate a pair of fabric denuders. However instead of an outlet to a pumping system, an open outlet is used. This open outlet was made from a section of machined 50mm PVC pipe. The other end of the PVC pipe is fitted with a similar pair of denuders, so there is symmetry on both sides of the
7 center axis through the PVC pipe. No flow restriction device was necessary. Sample collection
depends on the wind flowing through the assembly. The amount of ammonia collected, therefore,
is proportional to the wind speed, direction, and ammonia concentration. To measure the flux at a
point, two such samplers are needed: one on an east-west axis, the other on a north-south axis.
Figure 1 shows a schematic of this sampler in a north-south axis orientation.
An experiment was conducted to examine the flow abatement through the passive denuder assembly. A variable speed blower was used to simulate wind speeds from 2 to 7 m sec-
1. The laminar flow output from the blower was sufficiently large enough in diameter so as to encompass the front surface area (47 mm) of the denuder assembly. The simulated wind speed was measured with a Dwyer Wind Meter. The pressure drop inside between the pair of north-to- south or east-to-west fabric denuders was calibrated for different flows by using a manometer.
This allowed determining the flow versus pressure response of the denuder assembly. This could be related to the bulk air speed past the denuder. Measurements were then made of the pressure drop inside the denuder assembly for various air speeds generated by the blower. A plot of the two in Figure 2 shows that the flow through the denuder is linearly reduced by approximately a factor of 11. Since these tests were done under ideal conditions, where the air was blown directly on the front face of the denuder assembly and neglecting the turbulent and directional changes in field conditions, this can only be used as an approximation. Further work needs to be done to done to investigate the flow abatement through the denuder as a function not only of wind speed, but also of turbulence and wind direction.
8 Field Measurements
The California State University, Fresno, site is a medium-sized dairy operation (500 head). It
provided an opportunity to conduct controlled measurements and to study emissions from an
adjacent dairy lagoon. This also provided the best opportunity to compare the two types of
sampling techniques. The inclusion of emissions from both the dairy farm as well as the lagoon
was to eventually provide emission estimates for total dairy operation (to be reported elsewhere).
The investigation of the lagoon was scheduled to coincide with CIT’s (Center of Irrigation
Technology) investigation into lagoon acidification practices to investigate dairy lagoon
emissions from non-acidified lagoons, and acidified lagoons.
Six sets of passive denuders were used to collect samples at the dairy farm. Since the
prevailing winds are from the northwest, the sampling tower was selected to be approximately
100 meters downwind of the dairy with passive denuders at 1m, 2m (two at this elevation) and 5
m. The collocated sample set was used to estimate the overall precision of the measurement. An
upwind sample and another sample 100 meters downwind from the tower, both at 1m, completed
the six sample sets. A total of 48 denuders were required for each sampling day, with two days
sampled. The lagoon was located due north of the dairy and at a distance far enough from the dairy farm itself to be considered independent.
Four sets of passive denuders were used to sample the lagoon. Since the prevailing winds are
also from the northwest, the sampling tower was directly southeast of the primary lagoon. The samplers, both active and passive, as well as individual meteorological sensors were set at 1m,
2m and 5 m. An upwind sample at 1m completed the four sample sets. A total of 32 denuders were collected for each of three sampling days. To minimize flow variation through the passive
9 denuders, sampling was conducted during the time of day when winds were consistently from a
single direction. This was generally the case for the five sampling episodes as sampling times
were kept between three and four hours during the early afternoon when the winds were the most
stable. Two field blanks were taken for each sampling period. Table 2 lists the climatic
conditions during monitoring of NH3 emissions at the dairy lagoon.
Passive Denuder Analysis
The collected denuder samples were stored in airtight vials at 0°C and later analyzed in the
laboratory by the indophenol method29. The denuders were first extracted with deionized water.
Then reagents were added and the mixtures were allowed to react at room temperature for 45
minutes after which the absorbance at 635 nm of each sample was measured repeatedly using a
Beckman DU-600 spectrophotometer. Any sample exhibiting an absorbance greater than 1 was diluted and measured again. The calculation of the actual ammonia nitrogen concentrations in the
samples was done using a calibration curve based on an ammonium chloride standard solution.
To generate this calibration curve, six standard solutions of ammonium chloride with
concentrations in the range of 0–1 mg N/l were treated the same way as the denuder extracts. All
denuder ammonia measurements were corrected for field blank ammonium.
Table 3 shows the data from one of the experiment sets. The indophenol methods result in the
final concentration per denuder (micrograms per denuder). For the passive sampler, the
horizontal ammonia flux was determined by adding the net amount of ammonium that was
collected by each denuder and dividing this sum by the denuder’s collection surface area . From
the sample time (for this data set 240 min) and the active surface area of the denuder (13.9 cm-2)
the total flux through the denuder for each of the 4 directions was calculated. The strongly biased
directions for example the north and west for the dairy at one meter, show high denuder
10 efficiencies, greater than 90%. The directions opposite to the primary wind direction show much lower efficiencies and clearly have components of diffusion as well as turbulence. The lowest value of ammonia flux, which for this data set occurred for the dairy 5 m from the south, of 1.0 ng/cm2/min, can be used as a reasonable approximation for the diffusion component (D). Over the sampling days the diffusion component, determined this way, had a variance of 1.0 ng/cm2/min. Therefore the total net ammonia flux is the sum of the fluxes from the four directions (i) minus the diffusion component in each.
FNH3 = S(FNH3)i-4D (1)
Figures 3 and 4 are plots showing the contribution to the total ammonia flux from each direction for the two sampling periods at the dairy farm. Figure 4 is a plot of the data from Table
2. The dairy farm was located northwest to the sampling tower and winds came from the northwest with average speeds of 1.5 to 2.3 m/s. The average directions were between 295 to 305 degrees. The plots show that there is almost equal contribution to the total flux from the north and west denuders. There is a markedly strong gradient to the ground as the net flux contribution from the five-meter elevation is significantly lower. Since the wind speed, direction and temperature were fairly consistent for the two days, then the close agreement of the two data sets indicate that emission rates from the dairy are fairly consistent under similar climatic conditions.
RESULTS
The data from both the active and passive samplers were tabulated and are shown in Table 4.
Figure 5 shows a plot of the 15 data points scaled differently by a factor of 20. The trend is quite consistent between the two. By plotting the data against each other, the plot in Figure 6 shows that they are significantly correlated with an r2 of 0.773. The average difference between the two 11 is 19.9 with a standard deviation of 4.1.
Although this ratio exceeds the predicted laboratory results of approximately 11, many factors could bias the results of the passive sampler being low. First, at the lagoon most of the wind came from the northwest at approximately 290 to 300 degrees, and was not directly into the face of the denuders. If we assume just a cosine function on the wind direction, this would increase the ratio from 11 to approximately 15.5. Also, the passive denuders and active samplers were both sampled for a period of approximately 4 hours for each of the measurements. Since the flux from the active sampler represents an average of the concentration multiplied by an average of the wind speed, there are likely discrepancies from this as the passive sampler measures flux directly. There are also the effects of turbulence that need to be further investigated on the passive sampler. Finally, filter packs tend to be biased high since ammonium nitrate particulate collected on the filter are very likely to be partially volatilized.
To estimate the error for the passive samplers, two sets of passive samplers were collocated at the 2 m height for two of the sample measurements. The individual 16 fabric denuders from the primary denuder set used for the flux analysis were compared with the collocated set as shown in Figure 7. The individual denuders show a very strong correlation with r2=0.984. The offset of approximately 13% between the two is likely from inconsistencies in the phosphoric acid coating and sample analysis. The variance due to diffusive effects was found to
2 be approximately 1 ngNH3/cm /min, so that the data on Figures 4 and 5 for the passive samples
are FNH3 ± (.065(FNH3)+1). Active samples were tested and shown to have cumulative errors based on the flow measurement, wind speed measurement and inconsistencies of the citric acid coating and filter analysis. Overall there is no consideration to diffusive variance and the data on
Figure 4 for the active samples are FNH3 ± 0.08(FNH3).
12
The northwest bias of the individual denuders is observed from the plots in Figures 8 and
9. The stronger western component is due to the location of the tower and geometry of this particular dairy lagoon. The tower was located southeast of the secondary lagoon, but the primary lagoon was located due west of the secondary lagoon and therefore expected to increase the western component of the NH3 flux. Figures 8 and 10, pre and post acidification plots of the net ammonia flux, indicate that the net flux has been abated quite significantly by the acidification process. Before the addition of any acid, the average pH in the primary lagoon was
7.5, but it dropped to 6.3 on the day following the addition of approximately 1500 gallons of concentrated sulfuric acid. Overall, there was a decrease in the ammonia fluxes from the lagoon during the experiment, as shown in Figure 11. This is consistent with that of ammonia emission-
30,31 pH trends reported in the literature . A pH of greater than 7.5 in the lagoon favors NH3 volatilization18. It has also been reported that acidification of cattle slurry has been shown to
20 reduce NH3 emission rates .
+ Ammonia (NH3), which is in equilibrium with ammonium (NH 4) in the wastewater, is the end product of nitrogen degradation in the storage lagoon. The loss of NH3 to the atmosphere is in response to any differences in the NH3 partial pressures between that in equilibrium with the liquid phase and that in the ambient atmosphere. Hence the climatic and environmental factors that affected the NH3 partial pressure of the ambient atmosphere would have ultimately influenced the NH3 fluxes during the three days of the experiment. Table 2 shows the average values for wind speed, wind direction, air temperature, and relative humidity (RH), as well as the pH, temperature, and dissolved oxygen (DO) in the wastewater during the sampling at the lagoon.
13 In addition to the changes in pH of the wastewater in the lagoon, the minor climatic
changes in the parameters listed in Table 2 may also account for the differences in the ratio of
NH3 fluxes determined by the two methods observed during the three days of monitoring at the
lagoon. For example, the relatively highest NH3 flux observed at the 5 m location during the
acidification was probably influenced by the fact that the highest average wind speed was also
recorded at this location. Following the acid injection, the lagoon was aerated for approximately
two hours. The primary purpose for the aeration was to maintain a slightly aerobic layer. The
aerobic surface, evident by the slight increase in DO was expected to control the odors associated
with gases such as ammonia and hydrogen sulfide produced at the bottom of the anaerobic
32 lagoon . For example, under aerobic conditions the NH3, produced within the sub-surface
+ anaerobic layers, will first be converted to NH4 by heterotrophic bacteria and then into nitrate and nitrite by autotrophic bacteria. Hence, the nitrite and nitrate will remain solution and the amount of NH3 emitted from the lagoon will be reduced.
CONCLUSION
The passive flux denuder has shown to have the potential to be an accurate and economical method to measure ammonia fluxes from area sources. This information can be used with dispersion models to determine the overall emission rates for these types of sources. Additional research is needed to determine the cause of the discrepancy between the flux determined by the active and passive samplers.
14 Table 1. Phosphoric acid coated fabric denuder efficiency.
Nominal Flow Collection Conc. Temp RH Duration rate Denuder efficiency ppb C % days L/min Coating (w/w) %
NH3 38 21 17 0.9 2 2% phosphoric 66 NH3 38 20 20 1.8 2 9% phosphoric 90 NH3 40 38 20 2.7 2 9% phosphoric 95 NH3 15 21 20 1.1 2 9% phosphoric 100 NH3 15 21 80 7.0 2 9% phosphoric 100 NH3 13 38 22 4.9 2 9% phosphoric 100
15 Table 2. Climatic conditions during monitoring of NH3 emissions at the dairy lagoon.
Description Ambient Atmosphere Wastewater at Surface Height Speed Direction Temp. Relative pH Temp. Dissolved (m) (m/s) (degrees) (C) Humidity (%) (C) Oxygen (%)
Pre –acid 26.1 32.6 7.5 24.0 0.3 Pre-acid 1 1.12 292.9 Pre-acid 2 2.01 291.6 Pre-acid 5 2.10 297.0
During acid 25.9 35.9 6.9 23.9 0.3 During acid 1 1.83 300.7 During acid 2 2.91 297.1 During acid 5 3.09 303.2
Post acid 24.3 42.6 6.3 22.8 0.7 Post acid 1 0.94 240.6 Post acid 2 1.25 237.0 Post acid 5 1.12 242.9
16 Table 3. Set of dairy 2 passive denuder data showing total NH3 flux calculation.
Denuder Total Flux Total Sample Ammonia efficiency Through Ammonia ID Concentration Denuder Flux Field ug/denuder ng/cm2/min ng/cm2/min
4NF 111.6 94% 28.6 27.5 4NB 7.4 4SF 10.1 57% 4.2 3.2 4SB 7.6 4EF 4.1 38% 2.6 1.6 4EB 6.7 4WF 116.9 93% 30.2 29.2 4WB 8.8 DAIRY 1 METER 61.5
5NF 98.6 92% 25.6 24.6 5NB 8.2 5SF 10.2 52% 4.7 3.7 5SB 9.5 5EF 3.8 32% 2.9 1.8 5EB 8.1 5WF 106.9 93% 27.5 26.5 5WB 7.7 DAIRY 2 METER 56.6
6NF 112.6 94% 28.6 27.6 6NB 6.6 6SF 7.5 52% 3.4 2.4 6SB 6.9 6EF 6.3 36% 4.2 3.2 6EB 11.3 6WF 130.5 93% 33.7 32.7 6WB 9.8 DAIRY(Co-located) 2 METER 65.8
7NF 56.2 100% 13.5 12.4 7NB -0.1 7SF 3.0 69% 1.0 0.0 7SB 1.3 7EF 22.7 78% 7.0 5.9 7EB 6.3 7WF 87.2 99% 21.1 20.1 7WB 0.8 DAIRY 5 METER 38.5
17 Table 4. Comparison of active and passive denuder data.
Count Description Height Active Passive Ratio 2 2 (meters) (ngNH3/cm /min) (ngNH3/cm /min) 1 Pre-acid 1 991.7 52.5 18.9 2 Pre-acid 2 1026.4 59.1 17.4 3 Pre-acid 5 592.6 40.7 14.6 4 During acid 1 1031.2 70.2 14.7 5 During acid 2 946.9 63.2 15.0 6 During acid 5 804.1 42.3 19.0 7 Post acid 1 509.5 19.2 26.6 8 Post acid 2 391.5 15.1 25.9 9 Post acid 5 242.7 10.2 23.8 10 Dairy 1 1 1418.3 61.5 23.1 11 Dairy 1 2 1094.8 56.6 19.4 12 Dairy 1 5 839.2 38.5 21.8 13 Dairy 2 1 1798.9 82.3 21.9 14 Dairy 2 2 1282.2 85.1 15.1 15 Dairy 2 5 903.5 26.8 33.8
18 Figure 1. North-south passive sampler using 4 fabric denuders.
Figure 2. Flow abatement through the passive denuder assembly.
Figure 3. plot showing ammonia flux from the dairy versus wind direction (first day of measurements).
Figure 4. plot showing ammonia flux from the dairy versus wind direction (second day of measurements).
Figure 5. Comparison of active and passive denuder data for the 15 sample points.
Figure 6. Regression plot of active and passive denuder data.
Figure 7. Comparison of collocated passive denuder data.
Figure 8. plot showing ammonia flux from the dairy lagoon versus wind direction (pre- acidification).
Figure 9. plot showing ammonia flux from the dairy lagoon versus wind direction (during acidification).
Figure 10. plot showing ammonia flux from the dairy lagoon versus wind direction (post acidification).
Figure 11. Ammonia fluxes measured with active samplers before, during and after acidification of lagoon.
19 Figure 1. North-south passive sampler using 4 fabric denuders.
Denuder Substrates
Savillex Filter Spacers
NORTH SOUTH
1 1/2 inch PVC
Savillex PFA Nut Savillex PFA Extension
20
Figure 2. Comparison of air speed outside and inside the passive denuder assembly
8 y = 11.1x + 0.13 R2 = 0.997 6
4
Air Speed (m/s) 2
0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Air Speed Inside Denuder Assembly (m/s)
21 Figure 3. Plot showing ammonia flux from the dairy versus wind direction (first day of measurements).
North 45 40 35 NH3 Flux in 30 ng/cm2/sec 25 20 15 10 5 West 0 East
1m 2m 2m(Coloc) 5m
South
22 Figure 4. Plot showing ammonia flux from the dairy versus wind direction (second day of measurements).
North 45 40 35 NH3 Flux in 30 ng/cm2/sec 25 20 15 10 5 West 0 East
1m 2m 2m(Coloc) 5m
South
23 Figure 5. Comparison of active filter pack and passive denuder flux data for the 15 sample points.
2000 100.0
1800 90.0
Note that Mean Ratio Between Active and Passive Denuder Data is 19.9 1600 80.0
1400 70.0
1200 60.0
1000 50.0
800 40.0 Active samples' fluxes
600 30.0 Passive samples' fluxes
400 20.0
Samples 1 to 9 Are From the Dairy Lagoon Samples 10 to 15 Are From The Dairy Farm 200 10.0
0 0.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Measurement No.
Active Passive
24 Figure 6. Regression plot of active filter pack and passive denuder flux data.
100
90
80
70
60
50 y = 0.051x + 0.79 /min) (Passive Denuders) 2 R2 = 0.773 40 /cm 3 30
20 Flux (ngNH 3
NH 10 Note that the Intercept Almost Extends Through Zero (0.792) 0 0 200 400 600 800 1000 1200 1400 1600 1800 2000
2 NH3 Flux (ngNH3/cm /min) (Active Filter Packs)
25 Figure 7. Comparison of collocated passive denuder data.
50
45
NH3 Flux in 40 2 ngNH3/cm /min 35
30
25 y = 1.13x - 0.085 R2 = 0.984 Denuders 20
15
10 Flux from Co-located 2 meter Set of Passive 3 5 NH 0 0 5 10 15 20 25 30 35 40 45 50
NH3 Flux from Primary 2 Meter Set of Passive Denuders
26 Figure 8. Plot showing ammonia flux from the dairy lagoon versus wind direction (pre- acidification).
North 45 40 35 NH3 Flux in 30 ng/cm2/sec 25 20 15 10 5 West 0 East
1m 2m 5m
South
27 Figure 9. Plot showing ammonia flux from the dairy lagoon versus wind direction (during acidification).
North 45 40 35 NH3 Flux in 30 ng/cm2/sec 25 20 15 10 5 West 0 East
1m 2m 5m
South
28 Figure 10. Plot showing ammonia flux from the dairy lagoon versus wind direction (post acidification).
North 45 40 35 NH3 Flux in 30 ng/cm2/sec 25 20 15 10 5 West 0 East
1m 2m 5m
South
29 Figure 11. Ammonia fluxes measured with active samplers before, during and after acidification of lagoon.
200.0
180.0
160.0 /s) 2 140.0 /m
3 pH = 120.0 6.9
100.0 pH = 7.5 80.0
60.0 NH3 flux (m g NH 40.0 pH = 6.3 20.0 During Acidification Pre-Acidiification Post-Acidiification 0.0 0 1 2 3 4 5 6 Sample Elevation ( meters)
30 REFERENCES
1 Erisman, J.W.; Vermetten, A.W.M.; Asman, W.A.H.; Slanina, J.; and Wayers-Ijpelaan,
A. (1988) Vertical distribution of gases and aerosols: Behaviour of ammonia and related components in the lower atmosphere. Atmospheric Environment 22, 1153–1160.
2 Seinfeld, J.H. (1986) Atmospheric Chemistry and Physics of Air Pollution, Wiley, New
York.
3 Harrison, R.M. (1993) The chemistry and deposition of particulate nitrogen-containing species. In The Chemistry and Deposition of Nitrogen Species in the Troposphere, (edited by
Cocks, A.T.), pp. 95–105. The Royal Society of Chemistry, Cambridge.
4 Asman, W.A.H., and Janssen, A.J. (1987) A long-range transport model for ammonia and ammonium for Europe. Atmospheric Environment 21, 2099–2119.
5 Brost, R.A.; Delany, A.C.; and Huebert. B.J. (1988) Numerical modeling of concentrations and fluxes of HNO3, NH3, and NH4NO3 near the ground. Journal of Geophysical
Research 93 (D6), 7137–7152.
6 Finlayson-Pitts, B.J., and Pitts, J.N. (1986) Atmospheric Chemistry: Fundamentals and
Experimental Techniques, Wiley, New York.
7 Warneck, P. (1988) Chemistry of the Natural Atmosphere. Academic Press, New York.
31 8 Aneja, V.P.; Murray, G.; and Southerland, J. (1998) Atmospheric nitrogen compounds:
emissions, transport, transformation, deposition, and assessment. Environmental Manager, April,
22–25.
9 Heil, G.W, and Bruggink, M. (1987) Competition for nutrients between Calluna vulgaris
(l.) Hull and Molinia caerulea (L.) Moench. Oecologia 73, 105–107.
10 Pearl, H.W. (1995) Coastal eutrophication in relation to atmospheric nitrogen deposition: current perspectives. Ophelia 41, 237–259.
11 Sisler, J.F., and Malm, W.C. (1994) The relative importance of soluble aerosols to spatial and seasonal trends of impaired visibility in the United States. Atmospheric Environment 28,
851–862.
12 Bouwman, A.F.; Lee, D.S.; Asman, W.A.H.; Dentener, F.J.; van Der Hoek, K.W.; and
Oliver, J.G.J. (1997) A global high-resolution emission inventory for ammonia. Global
Biogeochemical Cycles 11(4), 561–587.
13 CDFA (1999) California Agricultural Resource Directory 1999. California Department of
Agriculture (CDFA), CA, USA.
14 Zhang, R. (1999) Integrated Animal Wastewater Treatment Systems. July 29th publication by the Biological and Agricultural Engineering Department, University of California at Davis.
32 15 Sheffield, R. (1998) Understanding Livestock Odors. A continuing education slide presentation from NC State University: http://www.bae.ncsu.edu/programs/extension/manure/ulo/.
16 Powlson, D.S. (1993) Understanding the soil nitrogen cycle. Soil Use and Management 9,
68–94.
17 Gaffney, P., and Shimp, D. (1999) Ammonia Emission Inventory Development: Needs,
Limitations, and What is Available Now. California Air Resources Board. Planning and
Technical Support Division, Sacramento, CA.
18 Zhang, R. (2001) Biology and engineering of animal wastewater lagoons. In Proc. of
California Plant and Soil conference. pp 71–79, California Chapter of American Society of
Agronomy and California Plant Health association, Feb 7–8, Fresno CA.
19 Walter, M.F.; Payne, V.W.E.; and Powers, T. (1999) Agricultural wastes and water, air and animal resources. Chapter 3 in Krider, J.N. and Rickman, J.D. (editors) Agriculture Waste
Management Field Handbook. NRCS publication available at: http://www.ftw.nrcs.usda.gov/awmfh/.html.
20 Bussink, D.W.; Huijsmans, J.F.M.; and Ketelaars, J.J.M.H. (1994) Ammonia volatilization from nitric-acid-treated cattle slurry surface applied to grassland. Netherlands J.
Agricultural Science 42, 293–309.
33 21 Ferm, M. (1975) Method for determination of atmospheric ammonia. Atmospheric
Environment 13, 1385–1393.
22 UCCE (2000) Using dairy lagoon water nutrients for crop production. A short course for
crop consultants conducted by University of California Cooperative Extension (UCCE) with
funding from USDA CSREES Water Quality Program.
23 Shaw, R.W., Jr.; Stevens, R.K.; Bowermaster, J.; Tesch, J.W.; and Tew, E. (1982)
Measurements of atmospheric nitrate and nitric acid: the denuder difference experiment.
Atmospheric Environment 16, 845–853.
24 Possanzini, M.; Febo, A.; and Liberti, A. (1983) A new design of a high-performance
denuder for the sampling of atmospheric pollutants. Atmospheric Environment 17, 2605–2610.
25 Koutrakis, P.; Sioutas, C.; Ferguson, S.T.; Wolfson, J.M.; Mulik, J.D.; and Burton, R.M.
(1993) Development and evaluation of a glass honeycomb denuder/filterpack system to collect atmospheric gases and particles. Environmental Science & Technology 27, 2437–2501.
26 Fitz, D.R., and Motallebi, N. (2000) A fabric denuder for sampling semi-volatile species.
J. Air & Waste Management Association 50, 981–992.
27 Fitz, D.R. and Tuazon, E.C. (2002) Evaluation of fabric denuders for the measurement of ammonia. in preparation.
34 28 Schjoerring, J.K. (1995) Long-term quantification of ammonia exchange between agricultural cropland and the atmosphere – I. Evaluation of a new method based on passive flux samplers in gradient configuration. Atmospheric Environment 29 (8), 885–893.
29 Scheiner, D. (1976) Determination of ammonia and Kjeldahl nitrogen by indophenol method. Water Research 10 (1), 31–36.
30 Freeney, J.R.; Simpson, J.R.; and Denmead, O.T. (1983) Volatilization of ammonia. In
Freeney, J.R. and Simpson, J.R. (editors) Gaseous Loss of Nitrogen from Plant-Soil Systems. pp
1–32. Martinus Nijhoff, The Hague.
31 Beauchamp, E.G.; Kidd, G.E.; and Thurtell, G. (1982) Ammonia volatilization from liquid dairy cattle manure in the field. Canadian Journal of Soil Science 62, 11–19.
32 Zhang, R.H.; Dugba, P.N.; and Bundy, D.S. (1997) Laboratory study of surface aeration of anaerobic lagoons for odor control of swine manure. American Society of Agricultural
Engineers 40 (1), 185–190.
ACKNOWLEDGMENTS
DISCLAIMER The statements and conclusions in this report are those of the authors and not necessarily those of the California Air Resources Board. The mention of commercial products, their source, or their
35 use in connection with material reported herein is not to be construed as actual or implied endorsement of such products.
ABOUT THE AUTHORS
36